The Pima Indians have the highest reported prevalence of NIDDM of any population in the world. Within this population, it is possible to identify subgroups of individuals at a particularly high risk for NIDDM. This project examines whether defects in insulin secretion contribute to the higher risk of NIDDM in these subgroups and whether they progress over the course of the disease. Healthy Pima men and women with normal glucose tolerance at high risk for NIDDM including individuals in the following 3 groups will be recruited: 1) persons whose mother and/or father developed diabetes at an early age (< 35 y); 2) persons whose mothers were diabetic during pregnancy; and 3) persons whose birth-weight was <2500 g. These individuals, as well as subjects with none of the above risk factors, will be admitted to the NIH Clinical Research Unit at Phoenix Indian Medical Center for the following series of studies. Body composition will be determined by DXA scanning and by measuring the amount of visceral abdominal fat using MRI. A 75-g oral glucose tolerance test will be performed. Insulin action will be measured with a hyperinsulinemic-euglycemic glucose clamp (insulin infusion: 40 mU/m2/min) and insulin secretory responses to glucose will be measured during a 5-step hyperglycemic glucose clamp immediately thereafter. This will be followed by a prolonged (42 hour) low-dose glucose infusion (4-6 mg/kg/min). A repeat euglycemic-step hyperglycemic glucose clamp will then be performed to test whether the glucose infusion alters insulin action and potentiates insulin secretion. Subjects will be followed longitudinally after discharge from the unit and oral glucose tolerance tests will be performed every three months. Individuals who transition from normal to impaired glucose tolerance or impaired to diabetic will be invited back to the Clinical Research Center for repeat testing. By comparing insulin secretion-glucose dose-response curves and the ability of glucose to potentiate insulin secretion in these groups, it may be possible to discern subtle defects in insulin secretion predisposing these individuals to NIDDM. In addition, comparison of the responses in the offspring of diabetic pregnancies with those in the offspring of mothers who subsequently became diabetic may allow us to separate defects due to genetic causes from those due to the intrauterine environment. To date, 12 subjects have completed the inititial testing phase. Preliminary data from these studies suggests that insulin secretory rates are a linear function of the level of hyperglycemia. Furthermore, there appear to be large differences between subjects in the slope of the curve relating insulin secretion to plasma glucose suggesting that this technique will be useful to distinguish subtle differences between groups of Pima Indians at high risk for NIDDM.